Structure-Based Drug Discovery​

Our research group is dedicated to advancing the field of computational drug discovery through the development and application of innovative hybrid methodologies that combine physics-based molecular modelling approaches with state-of-the-art artificial intelligence and machine learning techniques. By integrating these complementary paradigms, we aim to create robust, predictive, and scalable computational frameworks that address key challenges in modern drug discovery and accelerate the identification and optimization of novel therapeutic agents.

A major focus of our research is the development of computational strategies for structure-based drug discovery, leveraging detailed molecular and structural information to elucidate biomolecular recognition and guide rational therapeutic design. We apply a broad range of computational methods, including molecular docking, molecular dynamics simulations, free-energy calculations, quantum mechanical and hybrid QM/MM approaches, enhanced sampling techniques, and AI-driven predictive models. These methodologies enable us to investigate molecular interactions at multiple scales and provide mechanistic insights into protein function, ligand binding, and drug-target recognition.

Our computational platforms are employed to discover and optimize diverse classes of therapeutic agents, including small molecules, peptides, peptidomimetics, and other biologics. We focus on a wide range of protein targets whose dysregulation is associated with major human diseases, including cancer, infectious diseases, metabolic disorders, neurodegenerative conditions, and rare genetic diseases. Through the integration of structural biology, cheminformatics, molecular modelling, and data-driven approaches, we seek to identify novel chemical and biological entities with improved potency, selectivity, and therapeutic potential.

In addition to lead discovery, our group develops and applies advanced computational methods to support all stages of the drug development process. This includes the prediction and evaluation of absorption, distribution, metabolism, excretion, and toxicity (ADMET) properties, which are critical determinants of drug efficacy and safety. By combining molecular simulations, quantitative structure-activity relationship (QSAR) models, machine learning algorithms, and multi-parameter optimization strategies, we aim to enhance the early assessment of pharmacokinetic and toxicological profiles, thereby reducing experimental costs and improving candidate selection.

Beyond traditional drug discovery applications, we are also interested in developing next-generation computational tools that bridge mechanistic molecular modelling with modern AI techniques. These efforts seek to improve the accuracy, interpretability, and transferability of predictive models while enabling the efficient exploration of large chemical and biological spaces. Ultimately, our goal is to establish computational methodologies that not only facilitate the discovery of new therapeutics but also deepen our understanding of the molecular mechanisms underlying health and disease.

Through interdisciplinary collaborations spanning computational chemistry, structural biology, medicinal chemistry, bioinformatics, and data science, our research aims to contribute to the development of innovative therapeutic solutions and to advance the role of computational approaches as a central component of the modern drug discovery pipeline.